<<< Date Index >>>     <<< Thread Index >>>

Re: Problems with mutt and utf-8, can't talk to itself even!



On 19.03.08 14:37:58, Michael Kjorling wrote:
> On 19 Mar 2008 14:14 +0000, by cl@xxxxxxxx (Chris G):
> > Here are some incorrect pound signs:-
> > 
> >     ��������������������������������������������������������?�
> 
> This shows up to me (on a thoroughly UTF-8 system) as 56 undisplayable
> glyps, one question mark, and one more undisplayable glyph. Both
> gnome-terminal (cat and nano), gedit, and Emacs (in X) all give the
> same result. The binary data as stored in my mail spool is EF BF BD or
> U+FFFD "replacement character". According to the GNOME character map
> this is "used to replace an incoming character whose value is unknown
> or unrepresentable in Unicode".
> 
> So something is mangling your message, and my first guess at a suspect
> would be the editor still not understanding UTF-8 properly. Maybe
> someone can look at how this came through their end and confirm that
> suspicion, to rule out the possibility of my mail server doing
> something illicit, but if it is on my end, it's a first for me.

Exactly the same here, the binary data in the mail is not the pound
glyph

> > Here are some correct (as in correctly encoded as utf-8 by my editor)
> > pound signs:-
> > 
> >     ������
> 
> Same as above, except only six U+FFFD.

Yeap.

I've only had this when giving the editor some data which already was
encoded utf-8 and the editor thought its latin1 or some such and
re-encoded it (in my case I had a german signature from fortunes added
automatically to the template that mutt generates). So maybe the problem
really is that the file thats generated by mutt gets input from a
program that doesn't output utf-8 (or plain ascii) and the editor just
tries its best to somehow convert that stuff into valid utf-8.

Andreas

-- 
Caution: breathing may be hazardous to your health.